The 2010 revised WHO recommendations to provide antiretroviral (ARV) prophylaxis or treatment to mothers or infants during the breastfeeding period indicate a paradigm shift in PMTCT care and treatment programming. Yet despite South Africa's adoption of this guidance, myriad challenges currently exist. Confusion in the public health care system related to mixed messaging around safe infant feeding and the provision of-and now withdrawal of-free formula milk have made adherence to exclusive breastfeeding a challenge in South Africa. Cultural, social, and psychological factors influence the ability of women to follow PMTCT guidelines, which include exclusive breastfeeding for six months, adherence to ARV prophylaxis or treatment, and early infant diagnosis. Facility-based interventions alone are often inadequate to effect sustained behavioral changes in the face of multiple contextual factors. Community- and home-based support are needed, yet cost and systems constraints make these infeasible in many PMTCT programs. Our previous data suggest that a feeding buddy strategy could fill this gap and provide a home-based support system for the mother. The feeding buddy, who is selected by an HIV-positive pregnant woman to support her in overcoming sociocultural challenges to adhering to various aspects of PMTCT programs, is not an employed health care worker, but rather an individual known to the mother, making the intervention extremely cost-effective, and requiring minimal resources to implement. PATH is proposing a comprehensive evaluation of the feeding buddy concept in one health district of South Africa. The goal of the proposed study is to evaluate the effect of a feeding buddy to support mothers to adhere to PMTCT recommendations in order to establish feasible models of promoting HIV-free infant survival in resource- limited settings. We hypothesize that mothers who choose a feeding buddy will have increased rates of exclusive breastfeeding and adherence to ARV prophylaxis or treatment, as well as improved rates of early infant diagnosis and stigma reduction. A prospective cohort intervention study, set within a comprehensive ongoing national program addressing maternal and child health will be conducted with the following aims: (1) to determine the effect of a feeding buddy on adherence to exclusive breastfeeding and (2) to determine the effect of a feeding buddy on adherence to ARV prophylaxis or ART regimens. Secondary aims are: (1) to determine the effect of a feeding buddy on adherence to infant HIV testing at 6 weeks and (2) to determine the effect of a feeding buddy on disclosure and stigma. HIV-infected pregnant women (n=600) will be given the opportunity to choose a feeding buddy at an antenatal care visit to support infant feeding and PMTCT recommendations; follow-up will be to six months postpartum. Feeding buddies could be a simple, low-cost strategy for strengthening existing facility-level efforts to implement the new PMTCT guidelines, and ultimately could contribute toward improving HIV-free survival.